File size: 23,958 Bytes
d90395f |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 |
2023-10-17 08:23:49,889 ----------------------------------------------------------------------------------------------------
2023-10-17 08:23:49,890 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): ElectraModel(
(embeddings): ElectraEmbeddings(
(word_embeddings): Embedding(32001, 768)
(position_embeddings): Embedding(512, 768)
(token_type_embeddings): Embedding(2, 768)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): ElectraEncoder(
(layer): ModuleList(
(0-11): 12 x ElectraLayer(
(attention): ElectraAttention(
(self): ElectraSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): ElectraSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): ElectraIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): ElectraOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 08:23:49,890 ----------------------------------------------------------------------------------------------------
2023-10-17 08:23:49,891 MultiCorpus: 1100 train + 206 dev + 240 test sentences
- NER_HIPE_2022 Corpus: 1100 train + 206 dev + 240 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/de/with_doc_seperator
2023-10-17 08:23:49,891 ----------------------------------------------------------------------------------------------------
2023-10-17 08:23:49,891 Train: 1100 sentences
2023-10-17 08:23:49,891 (train_with_dev=False, train_with_test=False)
2023-10-17 08:23:49,891 ----------------------------------------------------------------------------------------------------
2023-10-17 08:23:49,891 Training Params:
2023-10-17 08:23:49,891 - learning_rate: "3e-05"
2023-10-17 08:23:49,891 - mini_batch_size: "8"
2023-10-17 08:23:49,891 - max_epochs: "10"
2023-10-17 08:23:49,891 - shuffle: "True"
2023-10-17 08:23:49,891 ----------------------------------------------------------------------------------------------------
2023-10-17 08:23:49,891 Plugins:
2023-10-17 08:23:49,891 - TensorboardLogger
2023-10-17 08:23:49,891 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 08:23:49,891 ----------------------------------------------------------------------------------------------------
2023-10-17 08:23:49,891 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 08:23:49,891 - metric: "('micro avg', 'f1-score')"
2023-10-17 08:23:49,891 ----------------------------------------------------------------------------------------------------
2023-10-17 08:23:49,891 Computation:
2023-10-17 08:23:49,891 - compute on device: cuda:0
2023-10-17 08:23:49,891 - embedding storage: none
2023-10-17 08:23:49,891 ----------------------------------------------------------------------------------------------------
2023-10-17 08:23:49,891 Model training base path: "hmbench-ajmc/de-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-17 08:23:49,891 ----------------------------------------------------------------------------------------------------
2023-10-17 08:23:49,891 ----------------------------------------------------------------------------------------------------
2023-10-17 08:23:49,892 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 08:23:50,594 epoch 1 - iter 13/138 - loss 3.49809368 - time (sec): 0.70 - samples/sec: 2981.66 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:23:51,325 epoch 1 - iter 26/138 - loss 3.16010129 - time (sec): 1.43 - samples/sec: 2869.08 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:23:52,079 epoch 1 - iter 39/138 - loss 2.75377183 - time (sec): 2.19 - samples/sec: 2827.93 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:23:52,821 epoch 1 - iter 52/138 - loss 2.28687909 - time (sec): 2.93 - samples/sec: 2854.62 - lr: 0.000011 - momentum: 0.000000
2023-10-17 08:23:53,636 epoch 1 - iter 65/138 - loss 1.95114714 - time (sec): 3.74 - samples/sec: 2856.95 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:23:54,405 epoch 1 - iter 78/138 - loss 1.72964478 - time (sec): 4.51 - samples/sec: 2894.38 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:23:55,133 epoch 1 - iter 91/138 - loss 1.58551835 - time (sec): 5.24 - samples/sec: 2887.28 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:23:55,867 epoch 1 - iter 104/138 - loss 1.44460998 - time (sec): 5.97 - samples/sec: 2872.53 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:23:56,599 epoch 1 - iter 117/138 - loss 1.33447301 - time (sec): 6.71 - samples/sec: 2878.42 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:23:57,359 epoch 1 - iter 130/138 - loss 1.24505589 - time (sec): 7.47 - samples/sec: 2871.22 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:23:57,831 ----------------------------------------------------------------------------------------------------
2023-10-17 08:23:57,832 EPOCH 1 done: loss 1.1903 - lr: 0.000028
2023-10-17 08:23:58,635 DEV : loss 0.25952714681625366 - f1-score (micro avg) 0.5734
2023-10-17 08:23:58,640 saving best model
2023-10-17 08:23:58,991 ----------------------------------------------------------------------------------------------------
2023-10-17 08:23:59,776 epoch 2 - iter 13/138 - loss 0.36122040 - time (sec): 0.78 - samples/sec: 2940.77 - lr: 0.000030 - momentum: 0.000000
2023-10-17 08:24:00,551 epoch 2 - iter 26/138 - loss 0.29793619 - time (sec): 1.56 - samples/sec: 2803.04 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:24:01,345 epoch 2 - iter 39/138 - loss 0.29683545 - time (sec): 2.35 - samples/sec: 2834.85 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:24:02,059 epoch 2 - iter 52/138 - loss 0.27965672 - time (sec): 3.07 - samples/sec: 2887.46 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:24:02,814 epoch 2 - iter 65/138 - loss 0.26522882 - time (sec): 3.82 - samples/sec: 2822.61 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:24:03,584 epoch 2 - iter 78/138 - loss 0.25845854 - time (sec): 4.59 - samples/sec: 2842.78 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:24:04,310 epoch 2 - iter 91/138 - loss 0.24532746 - time (sec): 5.32 - samples/sec: 2837.29 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:24:05,035 epoch 2 - iter 104/138 - loss 0.23679523 - time (sec): 6.04 - samples/sec: 2817.38 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:24:05,793 epoch 2 - iter 117/138 - loss 0.22779001 - time (sec): 6.80 - samples/sec: 2820.76 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:24:06,552 epoch 2 - iter 130/138 - loss 0.21987489 - time (sec): 7.56 - samples/sec: 2844.58 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:24:06,990 ----------------------------------------------------------------------------------------------------
2023-10-17 08:24:06,990 EPOCH 2 done: loss 0.2117 - lr: 0.000027
2023-10-17 08:24:07,641 DEV : loss 0.1423654556274414 - f1-score (micro avg) 0.821
2023-10-17 08:24:07,646 saving best model
2023-10-17 08:24:08,104 ----------------------------------------------------------------------------------------------------
2023-10-17 08:24:08,847 epoch 3 - iter 13/138 - loss 0.11206862 - time (sec): 0.74 - samples/sec: 2751.90 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:24:09,603 epoch 3 - iter 26/138 - loss 0.12075396 - time (sec): 1.50 - samples/sec: 2966.69 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:24:10,359 epoch 3 - iter 39/138 - loss 0.11031964 - time (sec): 2.25 - samples/sec: 2838.36 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:24:11,070 epoch 3 - iter 52/138 - loss 0.10673496 - time (sec): 2.96 - samples/sec: 2812.99 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:24:11,916 epoch 3 - iter 65/138 - loss 0.10361395 - time (sec): 3.81 - samples/sec: 2804.42 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:24:12,631 epoch 3 - iter 78/138 - loss 0.10060112 - time (sec): 4.53 - samples/sec: 2797.79 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:24:13,374 epoch 3 - iter 91/138 - loss 0.10082389 - time (sec): 5.27 - samples/sec: 2851.39 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:24:14,117 epoch 3 - iter 104/138 - loss 0.10284591 - time (sec): 6.01 - samples/sec: 2865.92 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:24:14,844 epoch 3 - iter 117/138 - loss 0.10302089 - time (sec): 6.74 - samples/sec: 2870.62 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:24:15,564 epoch 3 - iter 130/138 - loss 0.11214884 - time (sec): 7.46 - samples/sec: 2894.73 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:24:16,008 ----------------------------------------------------------------------------------------------------
2023-10-17 08:24:16,009 EPOCH 3 done: loss 0.1101 - lr: 0.000024
2023-10-17 08:24:16,647 DEV : loss 0.12610985338687897 - f1-score (micro avg) 0.8235
2023-10-17 08:24:16,652 saving best model
2023-10-17 08:24:17,087 ----------------------------------------------------------------------------------------------------
2023-10-17 08:24:17,830 epoch 4 - iter 13/138 - loss 0.05881840 - time (sec): 0.74 - samples/sec: 3208.55 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:24:18,531 epoch 4 - iter 26/138 - loss 0.06629479 - time (sec): 1.44 - samples/sec: 3022.27 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:24:19,259 epoch 4 - iter 39/138 - loss 0.06677456 - time (sec): 2.17 - samples/sec: 2991.15 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:24:20,005 epoch 4 - iter 52/138 - loss 0.06676999 - time (sec): 2.91 - samples/sec: 2924.52 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:24:20,771 epoch 4 - iter 65/138 - loss 0.07977619 - time (sec): 3.68 - samples/sec: 2966.27 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:24:21,479 epoch 4 - iter 78/138 - loss 0.07637301 - time (sec): 4.39 - samples/sec: 2938.14 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:24:22,202 epoch 4 - iter 91/138 - loss 0.07559118 - time (sec): 5.11 - samples/sec: 2939.35 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:24:22,960 epoch 4 - iter 104/138 - loss 0.07595665 - time (sec): 5.87 - samples/sec: 2941.79 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:24:23,715 epoch 4 - iter 117/138 - loss 0.07666906 - time (sec): 6.62 - samples/sec: 2936.63 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:24:24,440 epoch 4 - iter 130/138 - loss 0.07499893 - time (sec): 7.35 - samples/sec: 2916.68 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:24:24,933 ----------------------------------------------------------------------------------------------------
2023-10-17 08:24:24,934 EPOCH 4 done: loss 0.0750 - lr: 0.000020
2023-10-17 08:24:25,568 DEV : loss 0.1342494636774063 - f1-score (micro avg) 0.8682
2023-10-17 08:24:25,573 saving best model
2023-10-17 08:24:26,043 ----------------------------------------------------------------------------------------------------
2023-10-17 08:24:26,805 epoch 5 - iter 13/138 - loss 0.08072435 - time (sec): 0.76 - samples/sec: 2888.21 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:24:27,541 epoch 5 - iter 26/138 - loss 0.07670776 - time (sec): 1.50 - samples/sec: 2979.35 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:24:28,299 epoch 5 - iter 39/138 - loss 0.05993711 - time (sec): 2.25 - samples/sec: 2912.87 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:24:29,007 epoch 5 - iter 52/138 - loss 0.07266534 - time (sec): 2.96 - samples/sec: 2929.15 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:24:29,736 epoch 5 - iter 65/138 - loss 0.06792529 - time (sec): 3.69 - samples/sec: 2885.36 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:24:30,479 epoch 5 - iter 78/138 - loss 0.06388406 - time (sec): 4.43 - samples/sec: 2845.55 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:24:31,200 epoch 5 - iter 91/138 - loss 0.06078525 - time (sec): 5.15 - samples/sec: 2884.15 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:24:31,989 epoch 5 - iter 104/138 - loss 0.05854605 - time (sec): 5.94 - samples/sec: 2885.72 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:24:32,737 epoch 5 - iter 117/138 - loss 0.06512071 - time (sec): 6.69 - samples/sec: 2915.97 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:24:33,453 epoch 5 - iter 130/138 - loss 0.06281187 - time (sec): 7.41 - samples/sec: 2911.31 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:24:33,933 ----------------------------------------------------------------------------------------------------
2023-10-17 08:24:33,934 EPOCH 5 done: loss 0.0607 - lr: 0.000017
2023-10-17 08:24:34,581 DEV : loss 0.14152704179286957 - f1-score (micro avg) 0.8568
2023-10-17 08:24:34,586 ----------------------------------------------------------------------------------------------------
2023-10-17 08:24:35,308 epoch 6 - iter 13/138 - loss 0.02968530 - time (sec): 0.72 - samples/sec: 2875.23 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:24:36,102 epoch 6 - iter 26/138 - loss 0.04733392 - time (sec): 1.51 - samples/sec: 3078.82 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:24:36,808 epoch 6 - iter 39/138 - loss 0.04456611 - time (sec): 2.22 - samples/sec: 3059.39 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:24:37,515 epoch 6 - iter 52/138 - loss 0.04191343 - time (sec): 2.93 - samples/sec: 2998.65 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:24:38,264 epoch 6 - iter 65/138 - loss 0.03634236 - time (sec): 3.68 - samples/sec: 2963.01 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:24:39,043 epoch 6 - iter 78/138 - loss 0.03869706 - time (sec): 4.46 - samples/sec: 2941.49 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:24:39,777 epoch 6 - iter 91/138 - loss 0.04330849 - time (sec): 5.19 - samples/sec: 2932.89 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:24:40,523 epoch 6 - iter 104/138 - loss 0.04643481 - time (sec): 5.94 - samples/sec: 2908.17 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:24:41,234 epoch 6 - iter 117/138 - loss 0.04267826 - time (sec): 6.65 - samples/sec: 2903.73 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:24:41,988 epoch 6 - iter 130/138 - loss 0.04498568 - time (sec): 7.40 - samples/sec: 2899.92 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:24:42,430 ----------------------------------------------------------------------------------------------------
2023-10-17 08:24:42,430 EPOCH 6 done: loss 0.0511 - lr: 0.000014
2023-10-17 08:24:43,072 DEV : loss 0.15315352380275726 - f1-score (micro avg) 0.8651
2023-10-17 08:24:43,077 ----------------------------------------------------------------------------------------------------
2023-10-17 08:24:43,776 epoch 7 - iter 13/138 - loss 0.02477330 - time (sec): 0.70 - samples/sec: 2750.58 - lr: 0.000013 - momentum: 0.000000
2023-10-17 08:24:44,504 epoch 7 - iter 26/138 - loss 0.03433648 - time (sec): 1.43 - samples/sec: 2827.47 - lr: 0.000013 - momentum: 0.000000
2023-10-17 08:24:45,246 epoch 7 - iter 39/138 - loss 0.05028591 - time (sec): 2.17 - samples/sec: 2834.75 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:24:46,031 epoch 7 - iter 52/138 - loss 0.04693637 - time (sec): 2.95 - samples/sec: 2792.13 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:24:46,811 epoch 7 - iter 65/138 - loss 0.04956954 - time (sec): 3.73 - samples/sec: 2844.28 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:24:47,575 epoch 7 - iter 78/138 - loss 0.04284061 - time (sec): 4.50 - samples/sec: 2816.74 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:24:48,289 epoch 7 - iter 91/138 - loss 0.04173745 - time (sec): 5.21 - samples/sec: 2829.99 - lr: 0.000011 - momentum: 0.000000
2023-10-17 08:24:49,019 epoch 7 - iter 104/138 - loss 0.03971604 - time (sec): 5.94 - samples/sec: 2850.45 - lr: 0.000011 - momentum: 0.000000
2023-10-17 08:24:49,809 epoch 7 - iter 117/138 - loss 0.03842371 - time (sec): 6.73 - samples/sec: 2864.69 - lr: 0.000011 - momentum: 0.000000
2023-10-17 08:24:50,545 epoch 7 - iter 130/138 - loss 0.03787472 - time (sec): 7.47 - samples/sec: 2872.71 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:24:51,067 ----------------------------------------------------------------------------------------------------
2023-10-17 08:24:51,067 EPOCH 7 done: loss 0.0371 - lr: 0.000010
2023-10-17 08:24:51,704 DEV : loss 0.16822752356529236 - f1-score (micro avg) 0.8691
2023-10-17 08:24:51,708 saving best model
2023-10-17 08:24:52,153 ----------------------------------------------------------------------------------------------------
2023-10-17 08:24:52,907 epoch 8 - iter 13/138 - loss 0.03009041 - time (sec): 0.75 - samples/sec: 2792.32 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:24:53,660 epoch 8 - iter 26/138 - loss 0.04413822 - time (sec): 1.50 - samples/sec: 2898.69 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:24:54,389 epoch 8 - iter 39/138 - loss 0.04619894 - time (sec): 2.23 - samples/sec: 2929.53 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:24:55,144 epoch 8 - iter 52/138 - loss 0.04070750 - time (sec): 2.99 - samples/sec: 2919.08 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:24:55,937 epoch 8 - iter 65/138 - loss 0.04008572 - time (sec): 3.78 - samples/sec: 2898.93 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:24:56,665 epoch 8 - iter 78/138 - loss 0.03593628 - time (sec): 4.51 - samples/sec: 2871.59 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:24:57,405 epoch 8 - iter 91/138 - loss 0.03219739 - time (sec): 5.25 - samples/sec: 2903.86 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:24:58,149 epoch 8 - iter 104/138 - loss 0.03225186 - time (sec): 5.99 - samples/sec: 2879.86 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:24:58,877 epoch 8 - iter 117/138 - loss 0.03519688 - time (sec): 6.72 - samples/sec: 2894.29 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:24:59,631 epoch 8 - iter 130/138 - loss 0.03625943 - time (sec): 7.48 - samples/sec: 2893.82 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:25:00,043 ----------------------------------------------------------------------------------------------------
2023-10-17 08:25:00,043 EPOCH 8 done: loss 0.0352 - lr: 0.000007
2023-10-17 08:25:00,676 DEV : loss 0.18648800253868103 - f1-score (micro avg) 0.8666
2023-10-17 08:25:00,680 ----------------------------------------------------------------------------------------------------
2023-10-17 08:25:01,408 epoch 9 - iter 13/138 - loss 0.03893965 - time (sec): 0.73 - samples/sec: 2862.38 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:25:02,179 epoch 9 - iter 26/138 - loss 0.04375172 - time (sec): 1.50 - samples/sec: 2692.68 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:25:02,936 epoch 9 - iter 39/138 - loss 0.03344957 - time (sec): 2.25 - samples/sec: 2740.30 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:25:03,732 epoch 9 - iter 52/138 - loss 0.03076279 - time (sec): 3.05 - samples/sec: 2744.75 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:25:04,507 epoch 9 - iter 65/138 - loss 0.02960535 - time (sec): 3.83 - samples/sec: 2753.12 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:25:05,304 epoch 9 - iter 78/138 - loss 0.02789208 - time (sec): 4.62 - samples/sec: 2804.12 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:25:06,032 epoch 9 - iter 91/138 - loss 0.02960035 - time (sec): 5.35 - samples/sec: 2832.72 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:25:06,736 epoch 9 - iter 104/138 - loss 0.02787373 - time (sec): 6.05 - samples/sec: 2801.84 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:25:07,466 epoch 9 - iter 117/138 - loss 0.02791744 - time (sec): 6.78 - samples/sec: 2790.48 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:25:08,219 epoch 9 - iter 130/138 - loss 0.02832701 - time (sec): 7.54 - samples/sec: 2807.81 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:25:08,691 ----------------------------------------------------------------------------------------------------
2023-10-17 08:25:08,692 EPOCH 9 done: loss 0.0290 - lr: 0.000004
2023-10-17 08:25:09,498 DEV : loss 0.1811751425266266 - f1-score (micro avg) 0.873
2023-10-17 08:25:09,503 saving best model
2023-10-17 08:25:09,958 ----------------------------------------------------------------------------------------------------
2023-10-17 08:25:10,708 epoch 10 - iter 13/138 - loss 0.02736013 - time (sec): 0.75 - samples/sec: 2783.95 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:25:11,467 epoch 10 - iter 26/138 - loss 0.01588076 - time (sec): 1.51 - samples/sec: 2687.45 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:25:12,193 epoch 10 - iter 39/138 - loss 0.02320217 - time (sec): 2.23 - samples/sec: 2756.03 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:25:12,995 epoch 10 - iter 52/138 - loss 0.01910415 - time (sec): 3.03 - samples/sec: 2712.81 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:25:13,769 epoch 10 - iter 65/138 - loss 0.01657925 - time (sec): 3.81 - samples/sec: 2753.39 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:25:14,521 epoch 10 - iter 78/138 - loss 0.02142734 - time (sec): 4.56 - samples/sec: 2764.95 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:25:15,307 epoch 10 - iter 91/138 - loss 0.02070130 - time (sec): 5.35 - samples/sec: 2793.49 - lr: 0.000001 - momentum: 0.000000
2023-10-17 08:25:16,062 epoch 10 - iter 104/138 - loss 0.02375901 - time (sec): 6.10 - samples/sec: 2805.40 - lr: 0.000001 - momentum: 0.000000
2023-10-17 08:25:16,787 epoch 10 - iter 117/138 - loss 0.02537797 - time (sec): 6.83 - samples/sec: 2837.20 - lr: 0.000001 - momentum: 0.000000
2023-10-17 08:25:17,515 epoch 10 - iter 130/138 - loss 0.02350727 - time (sec): 7.55 - samples/sec: 2868.95 - lr: 0.000000 - momentum: 0.000000
2023-10-17 08:25:17,944 ----------------------------------------------------------------------------------------------------
2023-10-17 08:25:17,944 EPOCH 10 done: loss 0.0236 - lr: 0.000000
2023-10-17 08:25:18,599 DEV : loss 0.18580147624015808 - f1-score (micro avg) 0.8696
2023-10-17 08:25:18,935 ----------------------------------------------------------------------------------------------------
2023-10-17 08:25:18,936 Loading model from best epoch ...
2023-10-17 08:25:20,230 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-object, B-object, E-object, I-object, S-date, B-date, E-date, I-date
2023-10-17 08:25:20,863
Results:
- F-score (micro) 0.9096
- F-score (macro) 0.7442
- Accuracy 0.8505
By class:
precision recall f1-score support
scope 0.8977 0.8977 0.8977 176
pers 0.9839 0.9531 0.9683 128
work 0.8333 0.8784 0.8553 74
loc 1.0000 1.0000 1.0000 2
object 0.0000 0.0000 0.0000 2
micro avg 0.9108 0.9084 0.9096 382
macro avg 0.7430 0.7458 0.7442 382
weighted avg 0.9100 0.9084 0.9090 382
2023-10-17 08:25:20,863 ----------------------------------------------------------------------------------------------------
|